Author Topic: DrizzleIntegration Example  (Read 2956 times)

Offline Juan Conejero

  • PTeam Member
  • PixInsight Jedi Grand Master
  • ********
  • Posts: 6874
    • View Profile
    • http://pixinsight.com/
DrizzleIntegration Example
« on: 2014 May 28 13:30:56 »
Hi all,

The new DrizzleIntegration tool is almost ready for release. Just wanted to share with you the result of a test I've just run:


To the left, a normal integration of 47 undersampled wide-field images. To the right, the same images drizzled 2x with the new tool. Draw your own conclusions. This tool will be released in a few days.
Juan Conejero
PixInsight Development Team
http://pixinsight.com/

Offline Geoff

  • PixInsight Padawan
  • ****
  • Posts: 885
    • View Profile
Re: DrizzleIntegration Example
« Reply #1 on: 2014 May 28 14:24:29 »
There are clear improvements in the star shapes and contrast, but is the background somewhat noisier? This is just a visual impression. Really looking forward to the release.
Geoff
Don't panic! (Douglas Adams)
Astrobin page at http://www.astrobin.com/users/Geoff/
Webpage (under construction) http://geoffsastro.smugmug.com/

Offline mschuster

  • PTeam Member
  • PixInsight Jedi
  • *****
  • Posts: 1034
    • View Profile
Re: DrizzleIntegration Example
« Reply #2 on: 2014 May 28 14:59:45 »
Not sure what Juan implemented, but at least for classical drizzle there is a penalty of increased noise for improved resolution. On a per pixel basis, fewer subs contribute so per pixel SNR is lower. Another way to think about it is that the rejection ratio is higher. The resolution/noise tradeoff is fundamental.

Offline Ignacio

  • PixInsight Old Hand
  • ****
  • Posts: 375
    • View Profile
    • PampaSkies
Re: DrizzleIntegration Example
« Reply #3 on: 2014 May 29 13:09:21 »
The example looks very good, Juan. I am salivating already...

Ignacio

Offline Josh Lake

  • PixInsight Old Hand
  • ****
  • Posts: 424
    • View Profile
Re: DrizzleIntegration Example
« Reply #4 on: 2014 June 01 05:25:36 »
Not sure what Juan implemented, but at least for classical drizzle there is a penalty of increased noise for improved resolution. On a per pixel basis, fewer subs contribute so per pixel SNR is lower. Another way to think about it is that the rejection ratio is higher. The resolution/noise tradeoff is fundamental.

TGVDenoise to the rescue, eh?